How Important Is Optimal Gain Structure In A Modern PA System?

[BigRedX]

I ask because my band has recently had a number of quite slow soundchecks that appear to be caused by the PA engineer attempting to maximise the input gain on every channel.

We should be a fairly simple band to soundcheck - drums, sequenced synths and sound effects from a computer on a stereo pair of DIs. Two live synths each in mono on two more DIs, Bass Vi DI'd via a Line6 Helix and 1 vocal mic. We've worked hard on the relative levels of all our patches so it should simply be a question of pushing all 6 faders up and adjusting the level to get the vocals and drums nice and loud and the other instruments fitting into the mix.

However I've recently noticed a tendency for PA engineers to spend ages fiddling with the gain structure of each channel which is fine if you've got 30 minutes or so to soundcheck each band, but less good when there's two other bands waiting their turn and doors open in 20 minutes time.

While I can appreciate that getting the best signal to noise ratio is paramount in the studio, but live I've always been under the impression that so long as the channel isn't overloading at the input, and the faders aren't pushed right up to maximum to get an audible signal, setting the best possible gain structure isn't that important. You certainly shouldn't be spending an extra 5-10 minutes on it when time is scarce.

If I'm wrong please put me right.

[James Perrett]

I've not done serious live sound for a while but normally I'd solo each channel in turn and check that the signal level peaks up to around 0dBVU (NOT the same as 0dBFS) and adjust gain accordingly. If the signal is present then it should take no more than a few seconds for each channel.

However, nowadays there are all kinds of people involved in education and on YouTube who actually have very little proper experience at the coalface of live sound. Many of them were taught by teachers who had little real experience themselves. So they see things on video or read about things that make it appear as if they have to set everything perfectly - not realising that, when you are just one of 5 bands at the Bull & Gate, you are lucky if you get 10 minutes for a complete soundcheck.

[Drew Stephenson]

I've always taken an approximate approach (and 'under-gained') because, at the level of bands I've worked with, most of them will promptly turn their instruments up louder than the soundcheck as soon as they go on anyway.

[Wonks]

My last couple of experiences of live sound in a decent theatre have been pretty dire. The last one I couldn’t hear the bass player at all (except in a solo slot) and the kick was all very loud low resonant thud, totally unsuited to the type of music being played (it was so bad I left at the interval). I’ve heard some very well mixed performances there, so it’s not the venue that’s at fault.

So, as James says, I think a lot is down to a new bunch of soundpeople who really don’t know what they are doing. Perhaps down to the growing lack of smaller venues where they can learn their craft before venturing into the bigger venues.

[Hugh Robjohns]

As Drew and James say, PFL each channel in turn, set channel gain so the meters bounce on the safe side of 0VU* (anticipating a little more oomph when performing), maybe add any necessary EQ, job done. The rest is down to the faders.

It's not a once in a lifetime recording that needs to be perfect in every detail, it needs to be good enough, quickly.

*On a digital desk I'd be aiming at -20dBFS average level.

[Luke W]

As others have already said, this particular issue sounds like a lack of experience at the console end. If its audible with no noise and clipping, it'll probably do just fine!

My approach to rushed soundchecks was always to do a very rough pass at setting gains whilst kit was being rigged (find me a musician who doesn't start noodling as soon as they're setup and plugged in...), get everyone a monitor mix they're relatively happy with, and then ask them to play a song and do my fine-tuning/mix building whilst they do so.

It saved time, kept the band happy (nowhere near enough engineers put enough thought into this crucial step, in my humble opinion) and meant that I'd get a more accurate representation of how they were actually going to play.

My last couple of experiences of live sound in a decent theatre have been pretty dire. The last one I couldn’t hear the bass player at all (except in a solo slot) and the kick was all very loud low resonant thud, totally unsuited to the type of music being played (it was so bad I left at the interval).

Such a classic trick for so many live engineers... Spend too much time making the kick drum wobble the subs around and punch people in the chest, only to spend the rest of the evening trying to fight it armed with every other channel.

[Wonks]

With digital desks being the norm, and each channel having so much FX processing available, could it be that the more novice engineers feel they have to spend time adding compression and finessing reverbs on each channel rather than setting the gain, EQ if necessary and then moving on to the next channel to suit the time available?

[Murray B]

If it helps, I don't think you are wrong.

However, if the sound person is insistent that they have to use all the toys on the desk, setting the parameters on compressors and gates on everything can be a fiddly business - Perhaps this is the cause of the extra faff time.

Whether all this processing is entirely necessary for live sound...? I appreciate it on a big festival stage with a skilled engineer, I'm not sure if I need it playing in the back room of the Dog and Duck.

[James Perrett]

Whether all this processing is entirely necessary for live sound...? I appreciate it on a big festival stage with a skilled engineer, I'm not sure if I need it playing in the back room of the Dog and Duck.

The settings for the big festival stage will have been tuned in rehearsals in the weeks before the performance. All the sound engineer has to do is to load up the show file, playback a virtual soundcheck and fine tune a few things to suit the venue.

[Sam Spoons]

My approach at the small festival/quick change over gigs (or for my own pub gigs) was always to quickly set the individual gains and then ask the band to play a tune or two while I sorted out the balance. Fine detail like compression and 'special effects' (i.e. not reverb) was often redundant as we didn't have the time. But, on one gig, we had a visiting engineer with the headline band and, while his soundcheck was more like half an hour (and he knew the band intimately) rather than the 5 mins max we were allowed, his sound was definitely a step up from what I had* achieved.

But gain structure is a small part of the job, possibly the most important to get right before the show starts, but should be fairly straightforward so I suspect the tardy engineer you encountered was playing with the other toys which is all well and good but IME the audience get restless after a while.

Bear in mind that I was (and still occasionally do) working not so much at the grass roots of the live sound business but in the mud below them...

* This was back in the days before digital desks and we had provided a rack of outboard as specified in his rider with about 8 comps, a similar number of gates and a couple of extra delay/reverb processors in addition to the usual kit.

[Mike Monte]

I've always taken an approximate approach (and 'under-gained') because, at the level of bands I've worked with, most of them will promptly turn their instruments up louder than the soundcheck as soon as they go on anyway.

^^^^correct^^^^
Yes, all-of-them will play louder after sound check....including the singer.

Regarding drums: I usually listen to a drummer play for a moment and then match what the drum kit sounds like through the PA.
IMO: It should sound "the same", only louder (to fit the room).
Maybe I'm old (!)

[The Red Bladder]

You certainly shouldn't be spending an extra 5-10 minutes on it when time is scarce.
If I'm wrong please put me right.

You are dead right!

Long answer -

Sound guys (some even give themselves the title 'engineer' which is absurd!) often suffer from the delusion that they are important! They are there to get the sound right and not to faff about!

I once totted up the number of live gigs I had done in my live sound days - it came to about 2,000 - enough to allow me to pontificate on the subject.

In today's world of instant digital resets, the sound is set in rehearsals. The sound guy can do all the faffing about there and set the desk up to his/her heart's content. If in doubt, they can do a live recording of the band (or whatever it is that they are amplifying) and play with that without the band having to be there.

In my day (analogue days!) I marked the position for every knob with a Chinagraph pencil. The only variable was then the hall and a few were truly horrible, but most were OK-ish. Trick 17 was to fill the room with pink noise and watch what happens on switch-off on a fast acting real-time spectral analyser. The frequency that took the longest to die-down would be the culprit for any problems and I brought that down on the overall graphic EQ. In today's world, that task can be automated.

Best sound - Pink Floyd 'The Wall' at the Dortmund Westpfalenhalle. Sound by Britannia Row. The hall holds over 15,000 audience, so about three times the size of the Albert Hall, the speakers were mostly flown and the sound was in quad and crystal clear.

Worst sound - Prince at the Dortmund Westpfalenhalle. Sound (flown EV4 speakers) by Clair Bros. The usual mistake (at the orders of the artist of course!) was that the volumes were set so loud that the audience could even make out which song was being played. It wasn't distorting, they just overloaded the hall (and my head!)

Short answer -

The settings for the big festival stage will have been tuned in rehearsals in the weeks before the performance. All the sound engineer has to do is to load up the show file, playback a virtual soundcheck and fine tune a few things to suit the venue.

[Dave Rowles]

If he's spending ages fiddling with gain, then the system is setup completely wrong, or he's a very inexperienced engineer.

All my amps are set so that I can set my fader to zero mark, gain until I've got enough level, and then eq/compress/gate as needed. Takes me about 15/20 seconds a channel for initial setup, then I'll tweak through the first song.

If the system is set correctly and you're providing an consistent level, then it should take seconds to adjust all these settings. Let's face it, with digital these days, and knowing you've got the same mic package for each gig, a lot of the time a preset load plus slight gain adjustment and you're 90% there.

[shufflebeat]

There is another mistake that people make.

In very small set-ups you’re trying to match (with amplification) the sound of the quietest acoustic sources, vocals and acoustic melody instruments, to the loudest, drums and guitar amps. you are manipulating the sound of the system to fit the band.

At larger scale the acoustic sounds are a smaller element of the total so the character of the system is more apparent, and a powerful accelerant to ambition. This can encourage us to manipulate the band to show off the capacity of the system.

It’s so easy to lose control of the bottom end when the project wasn’t originally designed with that in mind.

With regard to the OP, well controlled initial gain structure gives us a much better platform to build with rather than problem-solving half way through the first song, annoying everyone.

[Dave Rowles]

Just to clarify...I meant tweak through the first song they play together in soundcheck...although a lot of tweaking can happen in the first song infront of the audience if there's been delays in soundchecks. That's not the ideal.

[AlecSp]

All my amps are set so that I can set my fader to zero mark, gain until I've got enough level, and then eq/compress/gate as needed.

With a system you know well and which is appropriately set up - this, a million times...

[tacitus]

Playing in a 10-piece function band we often get very little time to sound check, so one of us on horns has a fiddle in the opening track in case we goofed when setting levels. Since I keep a mixer just for the band, it’s not as horrifying as you’d think with an analogue desk.

Nearly always we match our volume to unmiked drums. Our drummer can play pretty loud and we have occasionally had trouble with sound limits. Besides, it’s not for me arbitrarily to ruin auntie Ethel’s hearing at a wedding. Though the late-night discos that often follow our stint usually aren’t bothered and turn everything up to 11.

[Sam Spoons]

DJ's are the sound guy's nemesis*, even when they're mates... I'm glad I don't have to deal with them anymore

I'm not talking about DJ/Producers doing something creative with decks and samples, rather 'party DJs' with, these days, a Spotify playlist consisting mainly of '80s hits and other such.

[shufflebeat]

Spotify playlist consisting mainly of '80s hits and other such.

There are few sights in this world more entertaining that a wedding DJ who doesn’t understand the concept of downloading files trying to link to the “free to customers” wifi internet connection.

[Sam Spoons]

Endless hours of entertainment...

[BigRedX]

Thanks everyone.

I think next time the PA engineer seems to be taking an inordinate amount of time playing with the channel settings on the desk, I'll go and have a look to see exactly what they are doing, and report back.

[LukaszPApassion]

Hey

My reply will be way too late and too long but hoping potentially useful every time the equipment comes to the 1st rehearsal of your gig.
Years ago I had an opportunity to jump on tour where MON engineer was a bit older than usually. Making story short, Ive put my IEMs on and been shocked straight away. Main artist mix was so amazing. So balanced, wide, natural, spacey and details everywhere Ive focused my ears into. Then, Ive started to look around and paused my eyes on the PSM1000 rack. I was observing PSM meters and for 90% of the song time meters were all green and occasionally first orange or down one green led below last green led. stood there and quickly had a thought: I am dealing with a bit older, previous generation Mix Engineer who would never call himself sound engineer on Facebook because he IS a proper soldier, no need to inform the whole world that he is a sound engineer. He just does his job the best he can.
Gain structure done with so much attention makes every piece of gear/module/chain/ all elements operating in the conditions that each of these segments is waiting for.
Gain structure done correctly along the board from AD Input to an output and further is the best move I can make to make this gig sounding better than expected.
Ive noticed that there is so many Facebook engineers that using their desk without knowing what meter type is in front of them at every channel. People using the RMS meter like it would be the Peak meter and vice versa.
Lets say Ive gained up my Beta91A on kick drum at +14dB what brings my meter to seat around line level for the most of the time and I have plenty of headroom in case drummer gets excited. To achieve his smile and my smile we both know that this is not a recording studio - this is live gig and we have 30 mins soundcheck and that's all. We can't change anything physically to get rid of boxy mids in this KickIn channel we can use EQ. so then We scoop very deep some lo-mids and even 1kHz wide, then we are closer to the goal and the most of the unaware Facebook engineers will go to the gate and compressor, smash it all up and "snare please"
there is one important thing that just happened - post EQ Ive lost a half of the level I had when I was coming into this EQ. There is no trim block post EQ like some old yamaha digital desks used to have. I must make my kick drum channel back to the line level I had so: I can use makeup to achieve that. Whatever flavour compressor I use , whatever attack-release time gets my transient undamaged and exposed so I have a chance. if the transient has gone because it looks cool, if the others can see my stylish warm vintage comp shining but the transient is gone- it will be better to get 1:1 ratio and use this comp for the most important reason right now-get my kick drum back to line level!
That's why I was amazed by Antons IEM mix with all meters everywhere are showing constant green and occasionally orange. Its because what was coming into the headamps was controlled so it could fit into preamp dynamic range and not collide with a drummer dynamic range that he brings to the show. It sounds simple but it can be difficult to achieve or totally forgotten by Facebook 'I am a sound engineer..I did this band, system was tuned by professionals so you can't change anything" etc.
We all have been on this path but some of us just been blessed by having a miracle and a chance to work with Anton like amazing person.
So all of the gates, comps, equalisers, de-essers all of those tools are smiling there inside the DSP because its a party time and the guy who turns the knob and fader is aware about gain structure and dynamic range.
So then delivering green occasionally orange to my stereo AUX, my IEM bus, my LR for one super important person that rely on me. He needs Vocal reverb, after snare room, snare plate, acoustic guitar plate- all of those tools that stop making an artist feel tired after 3mins of this show, getting all super dry direct and unnatural without creating some space around each instrument- these sends and returns also matching levels so I am hitting his plate aux in the way so its healthy occasionally orange level and using an AUX master EQ or de-esser to stop feeding my reverb engine by unwanted stuff which then would be multiplied later and pulling me to "I am not controlling this gig" situation.
So all my space design tools are using the same rule and I can always turn down return of plateFX rather than send less to it and land below healthy signal.
It helps with these stupid situations like: my channel is so hot that if I have to send on fader to somebody I don't have a full fader resolution and 1mm move does 10dB change etc. Just would make me fill like I shouldn't be there touching anything especially mixing board if I don't know how it all works.
To speed things up I step into the venue with my template file. doesn't matter if that's the desk file, PA system file, PA calc or anything. Just put some effort and build your own templates. It means that I am familiar with this equipment and if something goes down - I know how to use it because I have challenged myself to make my own from scratch.
Then we arrived at the AUX, Buss, MTX step and we have to do something that makes all green desk AUX = All green PSM transmiter or anything its patched to.
Getting to the device trim, read manual and set levels correctly so it will help me what's going on with PSM without constantly looking. Should say at the beginning do the same with RF mics-matching capsule gain for the voice is used for -bodypack to work with Fender Strat etc. Just pay attention before its getting amplified.
Mixing FOH its the same story.
If its too loud I turn my amps down respectively rather than pulling down PA matrix 9dB what ends up me loosing Bit Depth of my Master Buss and sig/noise ratio too, fader resolution again.
Just leave enough headroom for accents in your show and turn amps( hiss ) down
Make your own Smaart magnitude graph template so the PA can have enough headroom proportionally to last 17 gigs in theatres so you can see exactly what todays room has done vs splay vs Sub array level in this venue. All amps waiting for healthy line level too and behave if feed with care.
For me this was a game changer.
Apologies for my Polish-English mixture with almost all language rules destroyed without a shadow of awareness.

[Drew Stephenson]

Apologies for my Polish-English mixture with almost all language rules destroyed without a shadow of awareness.

Welcome Lukasz, lots of really good points in this and your other post.
And don't worry about your English, it's a very silly language...

[Arpangel]

I've always taken an approximate approach (and 'under-gained') because, at the level of bands I've worked with, most of them will promptly turn their instruments up louder than the soundcheck as soon as they go on anyway.

^^^^correct^^^^
Yes, all-of-them will play louder after sound check....including the singer.

Regarding drums: I usually listen to a drummer play for a moment and then match what the drum kit sounds like through the PA.
IMO: It should sound "the same", only louder (to fit the room).
Maybe I'm old (!)

How many times, I’ve lost count, when drums have been mic'ed up unnecessarily, or even needed them at all, especially pubs, which are normally quite small. They should balance to the acoustic sound of the kit.
I've assisted on quite large gigs in town halls, where we didn’t even bother to mic the kit, it just wasn’t necessary, plus, the players were experienced enough to more or less balance themselves, we had a decent PA, and sometimes got requests from vocalists about EQ etc, vocalists were normally the focal point, but once they were happy, it was just a minute or two on everything else, things like reverbs were never used much, just a bit of compression where needed.
After that it was getting the monitors sorted so they were all happy with their own levels, that’s what took the most time.

[Drew Stephenson]

The one thing I'd add about drums, is sometimes it can be difficult for the drummer to hear the kick properly (the acoustics of untreated rooms being what they are, particularly in a corner!), so they play it harder, which can lead to them playing everything harder.
On more than a couple of gigs I've mic'd the kick up and sent it to the drummer's monitor but not front of house. Drummer hears themselves better, doesn't feel the need to over-play. Brings down the level of everything else.